Cloud-Native Clinical Decision Support: Deploying Serverless Machine Learning Middleware for Real-Time Hospital Flow Optimization and Surgical Delay Prediction
The application of machine learning in healthcare presents unprecedented opportunities for optimizing hospital flow and mitigating surgical delays. However, the deployment of clinical decision support systems is frequently bottlenecked by the fragmented, unstructured nature of Electronic Health Records and the severe latency of legacy hospital IT infrastructure. This paper proposes a cloud-native architectural framework utilizing Amazon Web Services to construct a real-time, serverless pipeline for healthcare analytics. By deploying asynchronous Python middleware integrated with eXtreme Gradient Boosting algorithms, the proposed system programmatically ingests live patient admission data and historical morbidity metrics, instantly generating interpretable risk scores for surgical delays and patient prioritization. Preliminary architectural evaluations demonstrate that decoupling data ingestion from the predictive inference engine significantly reduces computational latency and ensures high availability during peak admission spikes. This methodology provides medical researchers and hospital administrators with a deterministic, highly scalable technological foundation for translating theoretical clinical AI into applied, real-world triage optimization.